Cytotoxic T-lymphocytes (cytotoxic T cells or CD8 T cells) involved in cell-mediated immunity monitor changes in cellular health by scanning peptide epitopes, or antigens, on cell surfaces. Peptide epitopes originate from cellular proteins and serve as a display mechanism that allows cells to present evidence of current cellular processes. Both native and non-native proteins (often referred to as self and non-self, respectively) are processed for peptide epitope presentation. Most self peptides are derived from natural protein turnover and defective ribosomal products. Non-self peptides may be derived from proteins produced in the course of events such as viral and bacterial infection, disease, and cancer.
Human tumors characteristically harbor a remarkable number of somatic mutations. In turn, expression of a peptide containing a mutation may be recognized as a non-self neoepitope by the adaptive immune system. Upon recognition of a non-self antigen, cytotoxic T cells will trigger an immune response resulting in apoptosis of cells displaying the non-self neoepitope. The cytotoxic T cell immune response is a highly specific mechanism of the adaptive immune system and is an efficient means for eliminating infected, diseased, and cancerous cells. There is a large therapeutic value in identifying immunogenic epitopes as exposure to immunogenic epitopes via vaccination may be used to trigger a desired cytotoxic T cell immune response. The role of immunogenic epitopes has been known in the scientific and medical community for decades, but the identification of antigens driving effective anti-tumor CD8 T cell responses remains largely unknown. The complexity involved with epitope presentation and cytotoxic T cell activation has mired their discovery and therapeutic use.
Major histocompatibility complex (MHC) class I molecules are responsible for peptide epitope presentation to cytotoxic T cells. In humans, the human leukocyte antigen (HLA) system is a locus of genes that code for MHC class I and class II molecules. HLA-A, -B, and -C genes code for MHC class I (MHCI) proteins. A peptide, typically 8-11 amino acids in length, will bind an MHCI molecule through interaction with a groove formed by two alpha helices positioned above an antiparallel beta sheet. Processing and presentation of peptide-MHC class I (pMHCI) molecules involve a series of sequential stages comprising: a) protease-mediated digestion of proteins; b) peptide transport into the endoplasmic reticulum (ER) mediated by the transporter associated with antigen processing (TAP); c) formation of pMHCI using newly synthesized MHCI molecules; and, d) transport of pMHCI to the cell surface.
On the cell surface, pMHCI will interact with cytotoxic T cells via T cell receptors (TCRs). Following the intricate pMHCI-TCR interaction, identification of a non-self antigen may result in cytotoxic T cell activation through a series of biochemical events mediated by associated enzymes, co-receptors, adaptor molecules, and transcription factors. An activated cytotoxic T cell will proliferate to produce a population of effector T cells expressing TCRs specific to the identified immunogenic peptide epitope. The amplification of T cells with TCR specificity to the identified non-self epitope results in immune-mediated apoptosis of cells displaying the activating non-self epitope.
The use of immunogenic epitopes to activate the immune system is currently being investigated for use in cancer therapy. While cancer cells originate from normal tissue, somatic mutations drive a large number of changes in the cancer proteasome. In turn, the resulting MHCI presented peptide epitopes, referred to as tumor-associated antigens (TAAs) or neoepitopes, allow for cytotoxic T cell differentiation between normal and cancer tissue. Recent work has confirmed that mutant peptides can serve as epitopes recognized as non-self by CD4 or CD8 T cells, but few mutant neoepitopes have been described since.
The use of peptide-based immunotherapy hinges on selection of a peptide epitope that will stimulate a desired cytotoxic T cell response. Specifically, tumor antigens can be classified into two categories: tumor-associated self-antigens (e.g., cancer-testis antigens, differentiation antigens) and antigens derived from shared or patient-specific mutant proteins. Since the presentation of self-antigens in the thymus may result in the elimination of high avidity T cells, mutant neoantigens are likely to be more immunogenic. The development of such immunotherapeutic epitopes is a challenging pursuit and efficient methods useful for the identification of efficacious epitopes are yet to be developed.
The time and cost intensive nature involved in the identification and verification of immunogenic peptide epitopes has hampered the development of efficacious peptide-based cancer vaccinations. To further complicate the issues involved in identifying immunogenic epitopes, permutations of mutations in cancer cells are often patient specific. The discovery of a mutant neoepitope requires laborious screening of a patient's tumor infiltrating lymphocytes for their ability to recognize an antigen from libraries constructed based on information from that patient's tumor exome sequence. Alternatively, mutant neoepitopes may be detected by mass spectrometry. However, mutant sequences have evaded detection because use of public proteomic databases that do not contain patient-specific mutations do not allow for their identification. The use of predictive algorithms, such as peptide-MHCI binding or peptide immunogenicity, may have potential application in the identification of personalized immunogenic epitopes. But, the vast number of somatic mutations and expression level changes contained in cancer cells results in a magnitude of predicted immunogenic epitopes too large for high-throughput immunogenic screening. Further, evidence of the poor immunogenicity of predicted epitopes calls into question the utility of current methodology.
There is need in the art to identify immunogenic epitopes suitable for use in peptide-based immunotherapy. Specifically, there is need in the art to identify immunogenic epitopes for use in peptide-based cancer therapy. Furthermore, there is need in the art for high-throughput methodology for prediction of immunogenic epitopes based on personalized genetic and/or proteomic analysis.
All references cited herein are hereby specifically incorporated by reference.